Chapter 7 Green 5G Femtocells for Supporting Indoor...
Transcript of Chapter 7 Green 5G Femtocells for Supporting Indoor...
Chapter 7
Green 5G Femtocells for Supporting Indoor
Generated IoT Traffic
Elias Yaacoub
Abstract Supporting the traffic emanating from the internet of things (IoT) is a ma-
jor challenge for 5G systems. A significant portion of this traffic will be generated
indoors. Therefore, in this chapter, femtocell networks designed for supporting IoT
traffic are studied. A deployment scenario of femtocell networks with centralized
control is investigated. It consists of an integrated wired/wireless system, where the
femtocell access points (FAPs) are controlled by a single entity. This permits per-
forming joint radio resource management in a centralized and controlled way in
order to enhance the quality of service performance for all users in the network. It
also allows an energy efficient operation of the network by switching off redundant
femtocells whenever possible. Two algorithms are proposed and analyzed. The first
one is a utility maximizing radio resource management algorithm, whereas the sec-
ond one is a FAP switch off algorithm, implemented at the central controller. The
joint wired/wireless resource management approach is compared to the distributed
resource management case, where each femtocell acts as an independent wireless
network unaware of the channel and interference conditions with the other cells.
The proposed algorithm was shown to lead to significant gains. Furthermore, con-
siderable energy savings were obtained with the green algorithm.
7.1 Introduction
One of the major challenges for 5G cellular systems is the capability to support
the machine-to-machine (M2M) traffic with the Internet of Things (IoT) becoming
a reality. In fact, IoT is expected to include billions of connected devices using
M2M communications [1]. These devices will have a variety of requirements and
different types of behavior in the network. For example, certain devices will access
Elias Yaacoub
Strategic Decisions Group (SDG) and Arab Open University (AOU), Beirut, Lebanon, e-mail:
1
2 Elias Yaacoub
the network frequently and periodically to transmit short amounts of data, such as
smart meters used for advanced metering infrastructure (AMI) in the smart grid [2,
3]. Other devices can store data measurements and transmit in bulk, unless there is
an alerting situation, such as sensor networks for environment monitoring [4]. In
fact, wireless sensor networks (WSNs) will constitute an integral part of the IoT
paradigm, spanning different application areas including environment, smart grid,
vehicular communication, and agriculture, among others [5]. Solutions to meet the
increasing demand include the deployment of heterogeneous networks involving
macrocells and small cells (picocells, femtocells, etc.), distributed antenna systems
(DAS), or relay stations (RSs).
A significant portion of IoT traffic will be generated indoors. This includes data
from smart meters (for electricity, water, etc.), from monitoring sensors (e.g., for
temperature, pollution levels inside an apartment or building, among other mea-
surements), and for home automation systems. IoT traffic can also emanate from
m-Health applications, with sensors relaying their monitoring data of elderly people
or indoor patients to the appropriate medical personnel and health centers [6]. Fem-
tocell Access Points (FAPs) can be used to handle this indoor traffic and reduce the
load on macrocell base stations (BSs). They generally consist of small, low power,
plug and play devices providing indoor wireless coverage to meet the quality of
service (QoS) requirements for indoor data users [7]. FAPs are installed inside the
home or office of a given subscriber. They are connected to the mobile operator’s
core network via wired links, e.g. digital subscriber line (DSL) [8]. However, they
are not under the direct control of the mobile operator since they are not connected
to neighboring macrocell BSs (MBSs) through the standardized interfaces, e.g., the
X2 interface for the long term evolution (LTE) cellular system.
This chapter investigates the case of 5G IoT in indoor scenarios, where the de-
ployment of FAPs is used to transmit the IoT traffic. Radio resource management
(RRM) algorithms are proposed for optimizing the resource allocation process and
meeting the quality of service (QoS) requirements of the IoT applications. Further-
more, techniques for the green operation of femtocell networks are proposed, with
the objective of maintaining QoS while ensuring an energy efficient operation of the
network.
The chapter is organized as follows. Femtocell networks are overviewed in Sec-
tion 7.2. The system model is presented in Section 7.3. The utility metrics leading to
different QoS and performance targets are described in Section 7.4. The joint RRM
algorithm implemented at the central controller is presented in Section 7.5, and the
FAP on/off switching algorithm is presented in Section 7.6. Simulation results are
presented and analyzed in Section 7.7. Finally, conclusions are drawn in Section 7.8.
7.2 Overview of Femtocell Networks
The proliferation of small cells, notably femtocells, is expected to increase in the
coming years [9]. Since most of the wireless traffic is initiated indoors, FAPs are
7 Green 5G Femtocells for Supporting Indoor Generated IoT Traffic 3
designed to handle this traffic and reduce the load on MBSs by providing indoor
wireless coverage to meet QoS requirements for indoor data users [7]. Since FAPs
are not under the direct control of the mobile operator and do not use the LTE X2
interface to connect to other BSs, they pose several challenges to network operation
and management.
A major challenge is that the overall interference levels in the network depend
on the density of small cells and their operation, which affects the configuration of
macrocell sites [10]. In [11], this problem was addressed by proposing macrocell-
femtocell cooperation, where a femtocell user may act as a relay for macrocell users,
and in return each cooperative macrocell user grants the femtocell user a fraction of
its superframe. In [12], it was assumed that both macrocells and small cells are
controlled by the same operator, and it was shown that in this case the operator can
control the system loads by tuning the pricing and the bandwidth allocation policy
between macrocells and small cells.
On the other hand, other works investigated radio resource management (RRM)
in femtocell networks by avoiding interference to/from macrocells. Most of these
works focused on using cognitive radio (CR) channel sensing techniques to deter-
mine channel availability. In [13], the femtocell uses cognitive radio to sense the
spectrum and detect macrocell transmissions to avoid interference. It then performs
radio resource management on the free channels. However, there is a time dedi-
cated for sensing the channel that cannot overlap with transmission/RRM time. A
channel sensing approach for improving the capacity of femtocell users in macro-
femto overlay networks is proposed in [14]. It is based on spatial radio resource
reuse based on the channel sensing outcomes. In [15], enhanced spectrum sensing
algorithms are proposed for femtocell networks in order to ensure better detection
accuracy of channels occupied by macrocell traffic.
In this chapter, LTE femtocell networks are investigated. FAPs are not assumed
to be controlled by the mobile operator. However, in certain scenarios, FAPs at a
given location can be controlled by a single entity. This can happen, for example, in
a university campus, hotel, housing complex, or office building. In such scenarios,
in addition to the wireless connection between FAPs and mobile terminals, FAPs
can be connected via a wired high-speed network to a central controller within the
building or campus. This can allow more efficient RRM decisions leading to signifi-
cant QoS enhancements for mobile users. Furthermore, it can allow energy efficient
operation of the network, by switching off unnecessary FAPs whenever possible,
and serving their active femto user equipment (FUEs) from other neighboring FAPs
that still can satisfy their QoS requirements. Due to centralized control, users do not
have to worry about opening the access to their FAPs for FUEs within the premises,
since the controller will guarantee the QoS. This scenario is studied in this chap-
ter, where two algorithms are presented: A utility maximizing RRM algorithm to
perform resource allocation over the FAPs controlled by the same entity, and an
algorithm for the green operation of LTE femtocell networks via on/off switching.
Significant gains are shown to be achieved under this integrated wired/wireless sce-
nario compared to the case where each FAP acts independently.
4 Elias Yaacoub
Fig. 7.1 Connections between a FAP and different IoT devices.
7.3 System Model
Fig. 7.1 shows multiple IoT devices that can be found indoors. They include smart
meters (for utilities: electricity, water, gas), sensors (for monitoring temperature,
humidity, environment parameters, or the performance of electrical appliances for
example), and body area networks (BANs) formed by sensors used to monitor a hu-
man being’s vital parameters for m-Health applications. The sensors of a BAN can
send this information via short range communications to the person’s mobile phone.
These IoT devices can communicate with the network through various technolo-
gies such as Bluetooth, Zigbee, or WiFi. In the case of BAN, the sensors actually
use these technologies to communicate with the patient’s smart phone, which in
turn communicates with the access point. With the deployment of 5G and the ex-
pected proliferation of IoT devices, these indoor devices can communicate with an
7 Green 5G Femtocells for Supporting Indoor Generated IoT Traffic 5
FAP
Network
Switch
Central
Controller
Fig. 7.2 System model.
indoor FAP using the cellular technology. With LTE-Advanced (LTE-A), this can
take place using device-to-device (D2D) communications for example. Similarly,
other devices such as laptops and mobile phones can still use the FAP normally, as
shown in Fig. 7.1. This allows these devices to benefit from advanced 5G features
guaranteeing QoS levels, and provides an integrated wireless network indoors with-
out incurring any additional costs, since the indoor communications between IoT
devices and the 5G FAP can be free of charge, similarly to Bluetooth or WiFi com-
munications. This comes at no loss for cellular operators too, since FAPs are user
installed devices and they relief the MBSs from this indoor generated traffic.
In this chapter, we consider a worst case scenario of a single device connected
to a FAP, and located at the opposite extremity from that FAP inside the house or
apartment. We also assume that the data rate requested by this device is equal to or
larger than the aggregate data rates of several IoT devices and other devices. For
example, real-time smart meter readings require a data rate of around 64 kbps [3],
whereas the data rates considered in the simulations of this chapter are orders of
magnitude larger (on the order of several Mbps). This allows simplifying the simu-
lations without losing the insights from the approach, since a worst case scenario is
adopted.
The system model of the worst case scenario with a single device per apartment,
denoted as the femto user equipment (FUE) is shown in Fig. 7.2. As an example,
a building having three apartments per floor is considered. One FAP is available in
each apartment, primarily to serve the FUEs available in that apartment. The FAPs
are connected to FUEs over the air interface, but they are connected via a wired
network (dashed lines in Fig. 7.2) to a central controller located within the building
(for example, in a room hosting telecom/networking equipment in the basement).
Interference is caused by the transmissions of a FAP to the FUEs served by the
other FAPs in other apartments. In the downlink (DL) direction from the FAPs to the
6 Elias Yaacoub
FAP
own
signal
DL
Interference
UL
Interference
Fig. 7.3 Intercell interference in the uplink and downlink.
FUEs, interference is caused by the transmissions of a FAP to the FUEs served by
the other FAPs in other apartments, as shown by the dashed lines in Fig. 7.3, repre-
senting the interference on the FUE in the second apartment in the third floor from
the FAPs in neighboring apartments. In the uplink (UL) direction from the FUEs
to the FAPs, interference is caused by the transmissions of an FUE to the FAPs in
other apartments, as shown by the dotted lines in Fig. 7.3, representing the interfer-
ence on the second FAP in the third floor from the FUEs in neighboring apartments.
Centralized RRM in an integrated wired/wireless scenario, as shown in Fig. 7.2, can
be used to mitigate the impact of interference and enhance QoS performance. In
addition, centralized control allows to switch certain FAPs off, or put them in sleep
mode, when they are not serving any FUEs, or when the FUEs they serve can be
handed over to other neighboring FAPs within the same building, without affecting
their QoS. An algorithm to implement this green switching approach is one of the
main contributions of this chapter, and is presented in Section 7.6.
In the absence of the central controller and wired connections between FAPs,
each FAP would act independently, without being aware of the network conditions
within the coverage areas of other FAPs. Thus, each FAP would selfishly serve its
own FUEs, regardless of the interference caused to other FUEs, or the redundant
energy consumption.
The building of Fig. 7.2 is assumed to be within the coverage area of an MBS,
positioned at a distance dBS from the building. The interference from the MBS to the
FAPs is taken into account in the analysis: it is assumed in this chapter that the MBS
is fully loaded, i.e. all its resource blocks (RBs) are occupied, which causes macro
interference to all the FAPs in the building. No coordination is assumed between the
mobile operator of the MBS and the central controller of the building FAPs.
7 Green 5G Femtocells for Supporting Indoor Generated IoT Traffic 7
Energy efficient FAP switching in conjunction with intelligent RRM is consid-
ered in this chapter within the framework of LTE. The downlink direction (DL)
from the FAPs to the FUEs is studied, although the presented approach can be easily
adapted to the uplink (UL) direction from the FUEs to the FAPs. In LTE, orthog-
onal frequency division multiple access (OFDMA) is the access scheme used for
DL communications. The spectrum is divided into RBs, with each RB consisting of
12 adjacent subcarriers. The assignment of an RB takes place every 1 ms, which is
the duration of one transmission time interval (TTI), or, equivalently, the duration
of two 0.5 ms slots [16, 17]. LTE allows bandwidth scalability, where a bandwidth
of 1.4, 3, 5, 10, 15, and 20 MHz corresponds to 6, 15, 25, 50, 75, and 100 RBs,
respectively [17]. In this chapter, scenarios where the MBS and the FAPs are using
the same bandwidth are assumed (i.e. a frequency reuse of one where bandwidth
chunks in different cells are not orthogonal).
7.3.1 Channel Model
The pathloss between FUE kl (connected to FAP l) and FAP j is given by [18]:
PLkl , j,dB =38.46+20log10 dkl , j +0.3dkl , j
+18.3n((n+2)/(n+1)−0.46)+qLiw
(7.1)
where dkl , j is the indoor distance between FUE kl and FAP j, n is the number of
floors separating FUE kl and FAP j, q is the number of walls between apartments,
and Liw is a per wall penetration loss. In (7.1), the first term 38.46+20log10 dkl , j is
the distance dependent free space path loss, the term 0.3dkl , j models indoor distance
dependent attenuation, the term 18.3n((n+2)/(n+1)−0.46) indicates losses due to prop-
agation across floors, and qLiw corresponds to losses across apartment walls in the
same floor. In this chapter, Liw = 5 dB is used as recommended in [18]. The pathloss
between FUE kl and its serving FAP l is a special case of (7.1), with j = l, n = 0,
and q = 0.
The FAPs in this chapter are assumed to be numbered from j = 1 to j = L, and
the outdoor MBS is represented by j = 0. The pathloss between FUE kl connected
to FAP l and the MBS j = 0 is given by [18]:
PLkl , j,dB = 15.3+37.6log10 dout,kl , j +0.3din,kl , j +qLiw +Low (7.2)
where dout,kl , j is the distance traveled outdoor between the MBS and the building
external wall, din,kl , j is the indoor traveled distance between the building wall and
FUE kl , and Low is an outdoor-indoor penetration loss (loss incurred by the out-
door signal to penetrate the building). It is set to Low = 20 dB [18]. In this chapter,
the MBS is considered to be located at a distance dBS from the building. Thus, the
indoor distance can be considered negligible compared to the outdoor distance. Fur-
thermore, the MBS is assumed to be facing the building of Fig. 7.3, such that q = 0
8 Elias Yaacoub
can be used. Thus, the outdoor-indoor propagation model of (7.2) becomes:
PLkl , j,dB = 15.3+37.6log10 dkl , j +Low (7.3)
Taking into account fading fluctuations in addition to pathloss, the channel gain
between FUE kl and FAP/MBS j can be expressed as:
Hkl ,i, j,dB =−PLkl , j,dB +ξkl , j +10log10 Fkl ,i, j (7.4)
where the first factor captures propagation loss, according to (7.1) or (7.2)-(7.3).
The second factor, ξkl , j, captures log-normal shadowing with zero-mean and a stan-
dard deviation σξ (set to σξ = 8 dB in this chapter), whereas the last factor, Fkl ,i, j,
corresponds to Rayleigh fading power between FUE kl and FAP or BS j over RB
i, with a Rayleigh parameter b such that E{|b|2} = 1. It should be noted that fast
Rayleigh fading is assumed to be approximately constant over the subcarriers of a
given RB, and independent identically distributed (iid) over RBs.
7.3.2 Calculation of the Data Rates
Letting Isub,kland IRB,kl
be the sets of subcarriers and RBs, respectively, allocated
to FUE kl in femtocell l, NRB the total number of RBs, L the number of FAPs, Kl the
number of FUEs connected to FAP l, Pi,l the power transmitted over subcarrier i by
FAP l, Pl,max the maximum transmission power of FAP l, and Rklthe achievable data
rate of FUE kl in femtocell l, then the OFDMA throughput of FUE kl in femtocell l
is given by:
Rkl(Pl,Isub,kl
) = ∑i∈Isub,kl
Bsub · log2(1+βγkl ,i,l) (7.5)
where Bsub is the subcarrier bandwidth expressed as Bsub = BNsub
, with B the total
usable bandwidth, and Nsub the total number of subcarriers. In (7.5), β refers to the
signal to noise ratio (SNR) gap. It indicates the difference between the SNR needed
to achieve a certain data transmission rate for a practical M-QAM (quadrature am-
plitude modulation) system and the theoretical limit (Shannon capacity) [19]. It
is given by β = −1.5ln(5Pb)
, where Pb denotes the target bit error rate (BER), set to
Pb = 10−6 in this chapter.
In addition, in (7.5), Pl represents a vector of the transmitted power on each
subcarrier by FAP/MBS l, Pi,l . In this chapter, the transmit power is considered to
be equally allocated over the subcarriers. Hence, for all i, we have Pi,l =Pl,max
Nsub.
The signal to interference plus noise ratio (SINR) of FUE kl over subcarrier i in
cell l in the DL, γkl ,i,l , is expressed as:
γkl ,i,l =Pi,lHkl ,i,l
Ii,kl+σ2
i,kl
(7.6)
7 Green 5G Femtocells for Supporting Indoor Generated IoT Traffic 9
where σ2i,kl
is the noise power over subcarrier i in the receiver of FUE kl , and Ii,klis
the interference on subcarrier i measured at the receiver of FUE kl . The expression
of the interference is given by:
Ii,kl=
L
∑j 6=l, j=0
(
K j
∑k j=1
αk j ,i, j
)
·Pi, jHkl ,i, j (7.7)
In (7.7), K j is the number of FUEs served by FAP j, and αk j ,i, j is a binary variable
representing the exclusivity of subcarrier allocation: αk j ,i, j = 1 if subcarrier i is
allocated to FUE k j in cell j, i.e., i ∈ Isub,k j, and αk j ,i, j = 0 otherwise. In fact,
in each cell, an LTE RB, along with the subcarriers constituting that RB, can be
allocated to a single FUE at a given TTI. Consequently, the following is verified in
each cell j:K j
∑k j=1
αk j ,i, j ≤ 1 (7.8)
The term corresponding to j = 0 in (7.7) represents the interference from the MBS,
whereas the terms corresponding to j = 1 to j = L represent the interference from
the other FAPs in the building.
7.4 Network Utility Maximization
In this section, the problem formulation for maximizing the network utility is pre-
sented. In addition, different utility metrics leading to different QoS objectives are
presented and discussed.
7.4.1 Problem Formulation
With U (l) and Ukldenoting the utility of FAP l and FUE kl , respectively, such that
U (l) = ∑Kl
kl=1 Ukl, then the objective is to maximize the total utility in the network of
Fig. 7.2, ∑Ll=1 U (l):
maxαkl ,i,l
,Pi,l
Utot =L
∑l=1
U (l) (7.9)
Subject to:Nsub
∑i=1
Pi,l ≤ Pl,max;∀l = 1, ...,L (7.10)
Kl
∑kl=1
αkl ,i,l ≤ 1;∀i = 1, ...,Nsub;∀l = 1, ...,L (7.11)
10 Elias Yaacoub
The constraint in (7.10) indicates that the transmit power cannot exceed the maxi-
mum FAP transmit power, whereas the constraint in (7.11) corresponds to the exclu-
sivity of subcarrier allocation in each femtocell, since in each LTE cell, a subcarrier
can be allocated at most to a unique user at a given scheduling instant. Different
utility functions depending on the FUEs’ data rates are described next.
7.4.2 Utility Selection
The utility metrics investigated include Max C/I, proportional fair (PF), and Max-
Min utilities. The impact of their implementation on the sum-rate, geometric mean,
maximum and minimum data rates in the network is studied in Section 7.7.1 using
the Algorithm of Section 7.5.
7.4.2.1 Max C/I Utility
Letting the utility equal to the data rate Uk = Rk, the formulation in (7.9) becomes
a greedy maximization of the sum-rate in the network. This approach is known
in the literature as Max C/I. However, in this case, FUEs with favorable channel
and interference conditions will be allocated most of the resources and will achieve
very high data rates, whereas FUEs suffering from higher propagation losses and/or
interference levels will be deprived from RBs and will have very low data rates.
7.4.2.2 Max-Min Utility
Due to the unfairness of Max C/I resource allocation, the need for more fair util-
ity metrics arises. Max-Min utilities are a family of utility functions attempting to
maximize the minimum data rate in the network, e.g., [20, 21]. A vector R of FUE
data rates is Max-Min fair if and only if, for each k, an increase in Rk leads to a
decrease in R j for some j with R j < Rk [20]. By increasing the priority of FUEs
having lower rates, Max-Min utilities lead to more fairness in the network. It was
shown that Max-Min fairness can be achieved by utilities of the form [21]:
Uk(Rk) =−R−a
k
a,a > 0 (7.12)
where the parameter a determines the degree of fairness. Max-Min fairness is at-
tained when a → ∞ [21]. We use a = 10 in this chapter. However, enhancing the
worst case performance could come at the expense of FUEs with good channel con-
ditions (and who could achieve high data rates) that will be unfavored by the RRM
algorithms in order to increase the rates of worst case FUEs. A tradeoff between
7 Green 5G Femtocells for Supporting Indoor Generated IoT Traffic 11
Max C/I and Max-Min RRM can be achieved through proportional fair (PF) utili-
ties, described next.
7.4.2.3 Proportional Fair Utility
A tradeoff between the maximization of the sum rate and the maximization of the
minimum rate could be the maximization of the geometric mean data rate. The ge-
ometric mean data rate for K FUEs is given by:
R(gm) =
(
K
∏k=1
Rk
)1/K
(7.13)
The metric (7.13) is fair, since an FUE with a data rate close to zero will make the
whole product in R(gm) go to zero. Hence, any RRM algorithm maximizing R(gm)
would avoid having any FUE with very low data rate. In addition, the metric (7.13)
will reasonably favor FUEs with good wireless channels (capable of achieving high
data rate), since a high data rate will contribute in increasing the product in (7.13).
To be able to write the geometric mean in a sum-utility form as in (7.9), it can
be noted that maximizing the geometric mean in (7.13) is equivalent to maximizing
the product, which is equivalent to maximizing the sum of logarithms:
maxK
∏k=1
Rk ⇐⇒maxln
(
K
∏k=1
Rk
)
= maxK
∑k=1
ln(Rk)
(7.14)
Consequently, the algorithmic implementation of (7.14) can be handled by the al-
gorithm of Section 7.5, by using, in that algorithm, Uk = ln(Rk) as the utility of
FUE k, where ln represents the natural logarithm. Maximizing the sum of loga-
rithms in (7.14) is equivalent to maximizing the product and is easier to implement
numerically. Hence, letting U = ln(R) provides proportional fairness [21, 22].
7.4.2.4 QoS-based Utility
The Max C/I, proportional fair (PF), and Max-Min utilities reflect the network per-
formance, but do not indicate if a specific FUE has achieved a desired QoS level or
not. For the green network operation, maximizing sum-rate or the minimum rate by
itself could prevent switching off certain FAPs. Instead, the objective in this case
would be to maximize the number of FUEs achieving their QoS requirements. Re-
sources allocated to increase the data rates beyond these requirements would be
redundant. Therefore, in this section, we propose a utility that reflects the number
of FUEs achieving a target data rate Rth, or how close they are to achieve it.
12 Elias Yaacoub
The utility function used for this purpose is expressed as follows:
Ukl= 1Rkl
≥Rth+1Rkl
<Rth
Rkl
Rth
(7.15)
In (7.15), the notation 1(Condition) is used such that 1(Condition) = 1 if condition is
verified, and 1(Condition) = 0 if the condition is not verified. This utility aims to max-
imize the number of FUEs who exceed their target data rate threshold Rth (first term
in (7.15)), or, if this is not achievable, reach a data rate as close as possible to Rth
(second term in (7.15), which corresponds to the fraction of Rth achieved by the
FUE). This utility is used with the Algorithm of Section 7.6 in order to obtain the
results of Section 7.7.2.
7.5 Centralized RRM Algorithm
To perform the maximization of (7.9), we use the utility maximization algorithm,
Algorithm 1, described in this section. This algorithm was first presented by the au-
thor in [23]. In this chapter, the energy efficiency aspects are added and investigated
through Algorithm 2 presented in Section 7.6, and the two algorithms are compared
in the results section. Algorithm 1 can be applied with a wide range of utility func-
tions, thus being able to achieve various objectives, with each objective represented
by a certain utility function. Hence, it can be used for max C/I, PF, and Max-Min
RRM, with the utilities derived in Section 7.4.2.
Lines 1-8 in Algorithm 1 are used for initialization. The loop in lines 10-21 deter-
mines the network utility enhancement that can be achieved by each (FUE, RB) al-
location. The allocation leading to maximum enhancement (Line 22) is performed if
it leads to an increase in network utility (Lines 23-30). After each allocation, the in-
terference levels in the network vary. Hence, interference and data rates are updated
and the novel utilities are computed. The process is repeated until no additional im-
provement can be obtained (Lines 9-34), with IImprovement being an indicator variable
tracking if an improvement in network utility has been achieved (IImprovement = 1) or
not (IImprovement = 0).
Algorithm 1 is implemented by the central controller in the scenario described in
Section 7.3. In this chapter, one FUE is considered to be active per femtocell, with-
out loss of generality. In the case where each FAP performs RRM in a distributed
way (without wired connections to a central controller), then the maximization of
the three utility types in each femtocell is achieved by allocating all the RBs of
a given FAP to the active FUE. In fact, in this case, there would be no informa-
tion about the channel gains and interference levels in the other femtocells. Thus,
it makes sense for each FAP to try to maximize the QoS of its served FUE by al-
locating all available resources to that FUE. For a given FAP l, this corresponds,
simultaneously, to maximizing the sum rate, maximizing the logarithm of the rate,
and maximizing the minimum rate (In fact, with one FUE kl present, Rklis the only
rate and thus would correspond to the sum rate, the minimum rate, and the geomet-
7 Green 5G Femtocells for Supporting Indoor Generated IoT Traffic 13
Algorithm 1 Utility Maximization Algorithm
1: for all FAP l and FUE kl do
2: for all RB j do
3: αoldkl , j
= 0
4: Uoldkl
(αold) = 0
5: end for
6: end for
7: Uoldtot =
L
∑l=1
Kl
∑kl=1
Uoldkl
(αold)
8: IImprovement = 1
9: while IImprovement = 1 do
10: for all FAP l and FUE kl do
11: for all RB j do
12: αnew = αold
13: αnewkl , j
= 1
14: for all FAP m and FUE km do
15: Calculate the interference and achievable data rates in the network
16: Calculate Unewkm
(αnew)17: end for
18: Unewtot =
L
∑l=1
Kl
∑kl=1
Unewkl
(αnew)
19: δkl , j =Unewtot −Uold
tot
20: end for
21: end for
22: Find (k∗, l∗, j∗) = argmaxk,l, j δkl , j
23: if δk∗l∗, j∗ > 0 then
24: αoldk∗
l∗, j∗ = 1
25: for all FAP m and FUE km do
26: Calculate the interference and achievable data rates in the network
27: Calculate Uoldkm
(αold)28: end for
29: Uoldtot =
L
∑l=1
Kl
∑kl=1
Uoldkl
(αold)
30: IImprovement = 1
31: else
32: IImprovement = 0
33: end if
34: end while
ric mean data rate in cell l). This uncoordinated allocation will lead to an increase
in interference levels, and to an overall degradation of performance in the network,
as shown by the results of Section 7.7.
It should be noted that Algorithm 1 allocates the resources of a given FAP exclu-
sively to the FUE served by that FAP, i.e., it supports closed access operation, al-
though it optimizes the performance by providing centralized control over the RRM
process. In a green networking scenario, certain FAPs can be switched-off and their
FUEs served by other FAPs in order to save energy. Hence, an algorithm with open-
14 Elias Yaacoub
Algorithm 2 RRM algorithm implemented at a given FAP l
1: for Nrounds = 1 to MaxRounds do
2: for l = 1 to L do
3: δl = 0
4: end for
5: Find l = argmin j;ξ j=1,δ j=0 ∑K j
k j=1 ∑NRBi=1 αk j ,i, j
6: kl = 0
7: δl = 1
8: HO OK = 1
9: while kl < Kl AND HO OK = 1 do
10: kl = kl +1
11: Find j∗ = argmax j;ξ j=1 ∑NRBi=1 γkl ,i, j
12: NAttempts = 0
13: Rkl= 0
14: while (Rkl< Rth) AND (∑
NRBi=1 αkl ,i, j
∗ < NRB) AND (NAttempts < MaxAttempts) do
15: NAttempts = NAttempts +1
16: Find i∗ = argmaxi;αkl ,i, j∗=0 γkl ,i, j
∗
17: Allocate RB i∗ to FUE kl : αkl ,i∗, j∗ = 1
18: Calculate the rate of FUE kl over RB i∗: Rkl ,i∗
19: Set Rkl= Rkl
+Rkl ,i∗
20: end while
21: if Rkl≥ Rth then
22: for all RB i such that αkl ,i,l = 1 do
23: αkl ,i,l = 0
24: end for
25: HO OK = 1
26: else
27: for all RB i such that αkl ,i, j∗ = 1 do
28: αkl ,i, j∗ = 0
29: end for
30: HO OK = 0
31: end if
32: end while
33: if HO OK = 1 AND ∑Kl
kl=1 ∑NRBi=1 αkl ,i,l = 0 then
34: ξl = 0
35: end if
36: end for
access operation, allowing FAP switch off while meeting the QoS requirements of
FUEs is required. Such an algorithm is presented in Section 7.6.
7.6 Green FAP Switching Algorithm
To perform centralized energy efficient operation of the femtocell network, the pro-
posed Algorithm 2, described in this section, is used. Algorithm 2 is implemented
by the central controller in the scenario described in Section 7.3. In this chapter, one
FUE is considered to be active per femtocell, without loss of generality, since Algo-
7 Green 5G Femtocells for Supporting Indoor Generated IoT Traffic 15
rithm 2 is applicable with any number of FUEs per femtocell. An FUE is considered
to be successfully served if it achieves a data rate above a defined threshold Rth.
In the algorithm, δl is a tracking parameter used to track if an attempt has been
made to switch off FAP l. It is set to δl = 1 if an attempt was made and to δl = 0
otherwise. ξl is a parameter indicating if FAP l is switched on or off. It is set to
ξl = 1 if the FAP is active and to ξl = 0 if it is switched off. In this chapter, we set
MaxRounds = L and MaxAttempts = NRB.
The algorithm finds the FAP that has the lowest load, with the load defined in
this chapter as the number of allocated RBs in the FAP (Line 5). It then makes an
attempt to switch off this FAP by moving its served FUEs to neighboring active
FAPs (Loop at Lines 9-32). The algorithm finds for each FUE, the best serving FAP
other than the current FAP l, in terms of best average SINR (Line 11). If the FUE can
be successfully handed over to the target FAP (and it can achieve its target rate after
resource allocation at Lines 14-20), it is handed over and the handover parameter
HO OK is set to 1 (Lines 21-25). If at least one FUE cannot be handed over, HO OK
is set to 0 and FAP l remains on after freeing any reserved RBs in the target FAP
(Lines 27-30). When all FUEs are handed over successfully, FAP l can be switched
off (Lines 33-35). Otherwise, if at least one FUE was not served successfully, FAP
l remains active.
7.7 Results and Discussion
6 RBs 15 RBs 25 RBs0
20
40
60
80
100
120
Sum
−R
ate
(M
bps)
6 RBs 15 RBs 25 RBs0
10
20
30
40
Geom
. M
ean D
ata
Rate
(M
bps)
6 RBs 15 RBs 25 RBs0
20
40
60
80
Max.
Rate
(M
bps)
6 RBs 15 RBs 25 RBs0
5
10
15
20
25
30
Min
. R
ate
(M
bps)
Max C/I
PF
Max−Min
Dist.
Fig. 7.4 Results in the case of one floor (three femtocells).
This section presents the Matlab simulation results obtained by implementing the
proposed approach under the system model of Section 7.3. We consider a building as
shown in Fig. 7.2. Three apartments per floor are assumed, with one active FUE per
16 Elias Yaacoub
6 RBs 15 RBs 25 RBs0
50
100
150
Su
m−
Ra
te (
Mb
ps)
6 RBs 15 RBs 25 RBs0
5
10
15
20
25
Ge
om
. M
ea
n D
ata
Ra
te (
Mb
ps)
6 RBs 15 RBs 25 RBs0
20
40
60
80
Ma
x.
Ra
te (
Mb
ps)
6 RBs 15 RBs 25 RBs0
5
10
15
20
Min
. R
ate
(M
bp
s)
Max C/I
PF
Max−Min
Dist.
Fig. 7.5 Results in the case of two floors (six femtocells).
6 RBs 15 RBs 25 RBs0
50
100
150
200
Sum
−R
ate
(M
bps)
6 RBs 15 RBs 25 RBs0
5
10
15
20
Geom
. M
ean D
ata
Rate
(M
bps)
6 RBs 15 RBs 25 RBs0
20
40
60
80
Ma
x.
Rate
(M
bps)
6 RBs 15 RBs 25 RBs0
5
10
15
Min
. R
ate
(M
bps)
Max C/I
PF
Max−Min
Dist.
Fig. 7.6 Results in the case of three floors (nine femtocells).
apartment using the FAP to access the network (assuming one FAP per apartment).
The maximum FAP transmit power is set to 1 Watt, whereas the transmit power of
the macro BS is set to 10 Watts.
7.7.1 Results of Centralized RRM with all the FAPs Active
This section presents the results of implementing Algorithm 1 described in Sec-
tion 7.5 when all the FAPs are active. Scenarios with one floor only (three apart-
ments on ground floor), two floors (six apartments), and three floors (nine apart-
7 Green 5G Femtocells for Supporting Indoor Generated IoT Traffic 17
ments) are investigated, with the results shown in Figs. 7.4, 7.5, and 7.6, respec-
tively.
The figures show that max C/I scheduling leads to the highest sum-rate in the
network. However, this comes at the expense of fairness, as it can be seen from the
geometric mean results of max C/I. In fact, the bottom subfigures of Figs. 7.4 to 7.6
show that max C/I enhances the maximum rate in the network, by allocating most of
the resources to the FUE having the best channel and interference conditions, while
depriving other FUEs from sufficient resources, thus leading to unfairness, as shown
by the minimum rate plots. On the other hand, PF scheduling maximizes the geo-
metric mean for all the investigated scenarios. Clearly, the minimum rates achieved
with PF indicate that a PF utility is significantly more fair than max C/I. The results
of Max-Min scheduling also show a fair performance. In fact, Max-Min resource
allocation leads to maximizing the minimum rate in the network for almost all the
studied scenarios, except in the case of one and two floors with six RBs, where it is
slightly outperformed by PF. This is due to the approximation performed by taking,
in (7.12), a = 10 instead of a = ∞. When the number of resources increases to 15
and 25 RBs, the algorithm has additional flexibility to implement RRM with Max-
Min such that the minimum rate is maximized compared to the other methods. It can
also be noted that Max-Min scheduling leads to a geometric mean performance that
is reasonably close to that of PF scheduling, indicating that it also enhances overall
fairness in the network. Figs. 7.4 to 7.6 also show that, as expected, the data rates
increase for all the studied metrics when the number of RBs increases.
Comparing the joint wired/wireless case to the distributed scenario where each
FAP performs RRM independently without centralized control, it can be seen that
the distributed scenario is outperformed by the integrated wired/wireless approach
for all the investigated metrics: Max C/I leads to a higher sum-rate, PF leads to a
higher geometric mean, and Max-Min leads to a higher minimum rate. This is due to
the fact that with distributed RRM, a FAP is not aware of the interference conditions
to/from other FAPs and FUEs. This leads to a severe performance degradation, as
can be seen in Figs. 7.4 to 7.6, although all the RBs of a given FAP are allocated to
the FUE served by that FAP.
7.7.2 Results of the Green Network Operation with FAP On/Off
Switching
This section presents the simulation results, considering the scenario of Fig. 7.2,
with different values for Rth and the available LTE bandwidth. The following meth-
ods are compared:
• The centralized scheduling algorithm presented in Section 7.5. It assumes each
FAP serves only its corresponding FUEs without taking energy efficiency into
account. But the resource allocation is performed by the central controller, which
allows to avoid interference.
18 Elias Yaacoub
Table 7.1 Average Data Rates (Mbps)
Centralized Green Selfish
Centralized
NRB = 15
Rth = 2 Mbps 2.74 3.01 5.25
NRB = 15
Rth = 5 Mbps 6.01 6.06 5.25
NRB = 15
Rth = 7 Mbps 7.64 7.18 5.25
NRB = 15
Rth = 10 Mbps 9.44 8.94 5.25
NRB = 25
Rth = 5 Mbps 6.09 6.14 8.54
NRB = 25
Rth = 7 Mbps 7.82 7.87 8.54
NRB = 25
Rth = 10 Mbps 10.94 10.91 8.54
NRB = 50
Rth = 10 Mbps 11.00 11.04 15.90
• The “selfish” approach, where each FBS allocates all its RBs to the FUE it is
serving, regardless of the allocations in other cells. This scenario assumes neither
centralized control, nor any form of coordination between FAPs. Thus, it would
be logical for each FAP to allocate all resources to its served FUEs, given that no
other coordination or interference information is available.
• The approach proposed in Algorithm 2, where, starting from an initial allocation
without energy efficiency obtained by implementing Algorithm 1, the proposed
Algorithm 2 implements centralized FAP switching off after offloading FUEs to
active FAPs that can maintain their QoS.
In this section, we use a capped capacity formula in order to limit the possibility
of FUEs to achieve their target rate:
Rkl(Pl,Isub,kl
) =
max(
∑i∈Isub,kl
Bsub · log2(1+βγkl ,i,l),Rmax
)
(7.16)
Compared to (7.5), the expression in (7.16) is a capped Shannon formula; i.e., the
data rate is not allowed to exceed the maximum limit Rmax that can be reached using
7 Green 5G Femtocells for Supporting Indoor Generated IoT Traffic 19
Table 7.2 Geometric Mean Data Rates (Mbps)
Centralized Green Selfish
Centralized
NRB = 15
Rth = 2 Mbps 2.15 2.94 2.83
NRB = 15
Rth = 5 Mbps 5.83 6.00 2.83
NRB = 15
Rth = 7 Mbps 7.46 6.57 2.83
NRB = 15
Rth = 10 Mbps 8.84 6.01 2.83
NRB = 25
Rth = 5 Mbps 5.92 6.07 4.66
NRB = 25
Rth = 7 Mbps 7.70 7.83 4.66
NRB = 25
Rth = 10 Mbps 10.84 10.78 4.66
NRB = 50
Rth = 10 Mbps 10.91 11.01 8.56
practical modulation and coding schemes (MCS) in LTE. This limit is determined
as follows:
Rmax =rn ·N
(kl)RB ·N
RBSC ·N
SCSymb ·N
TTISlot
TTTI, (7.17)
where rn is the rate in bits/symbol corresponding to the MCS used over the subcar-
riers of the RBs allocated to the FUE. Rmax is obtained with rn = 6 corresponding to
uncoded 64-QAM, the highest MCS used in LTE. In addition, N(kl)RB is the number of
RBs allocated to kl , NRBSC is the number of subcarriers per RB (equal to 12 in LTE),
NSCSymb is the number of symbols per subcarrier during one time slot (set to six or
seven in LTE, depending whether an extended cyclic prefix is used or not), NTTISlot is
the number of time slots per TTI (two 0.5ms time slots per TTI in LTE), and TTTI is
the duration of one TTI (1ms in LTE) [16].
We use the utility (7.15) with both Algorithms 1 and 2. The average data rate
results are shown in Table 7.1. However, the average rate results alone can be mis-
leading. In fact, when an FUE A has a very high data rate while another FUE B has a
very poor data rate, the average might still be high, but the poor performance of FUE
B is masked by the high rate of FUE A. Using the geometric mean results provides
a better indication of fairness. The geometric mean data rate results are presented in
20 Elias Yaacoub
Table 7.3 Fraction of FUEs in Outage
Centralized Green Selfish
Centralized
NRB = 15
Rth = 2 Mbps 0.16 0.0 0.36
NRB = 15
Rth = 5 Mbps 0.11 0.0 0.62
NRB = 15
Rth = 7 Mbps 0.15 0.13 0.73
NRB = 15
Rth = 10 Mbps 0.55 0.44 0.83
NRB = 25
Rth = 5 Mbps 0.11 0 0.46
NRB = 25
Rth = 7 Mbps 0.11 0 0.57
NRB = 25
Rth = 10 Mbps 0.10 0.01 0.68
NRB = 50
Rth = 10 Mbps 0.11 0 0.49
Table 7.2. In addition, Table 7.3 shows the fraction of FUEs in outage, i.e. the num-
ber of FUEs that did not achieve Rth divided by the total number of FUEs. Table 7.4
shows the fraction of FAPs that are active in order to serve the FUEs. Naturally, the
centralized and selfish cases have all their values equal to 1, since 100% of the FAPs
are active. Table 7.5 shows the value of the utility function (7.15).
Tables 7.1-7.5 show that the centralized scheduling approach and the centralized
green approach significantly outperform the selfish method, especially in terms of
fairness and outage. The results of Table 7.4 indicate that the proposed green method
of Algorithm 2 is achieving significant energy savings, as it is using only one or two
FAPs to serve the nine FUEs (indeed, the value 0.11 corresponds to the ratio 1/9).
This is an interesting result, since it indicates that FUEs in neighboring apartments
can be successfully served by a single FAP, which saves around 90 % of FAP energy
consumption.
Comparing the results of Algorithm 1 to Algorithm 2, Tables 7.1, 7.2, and 7.5
show that they have a comparable performance, with one being slightly better than
the other, or vice versa. However, interestingly, Table 7.3 shows that Algorithm 2 al-
ways leads to better outage performance. This is explained by the fact, that, although
fully centralized and using all FAPs, Algorithm 1 operates under the constraint that
a FAP serves only the FUEs in its apartment. Hence, although centralized control
7 Green 5G Femtocells for Supporting Indoor Generated IoT Traffic 21
Table 7.4 Fraction of Active FAPs
Centralized Green Selfish
Centralized
NRB = 15
Rth = 2 Mbps 1 0.11 1
NRB = 15
Rth = 5 Mbps 1 0.11 1
NRB = 15
Rth = 7 Mbps 1 0.12 1
NRB = 15
Rth = 10 Mbps 1 0.21 1
NRB = 25
Rth = 5 Mbps 1 0.11 1
NRB = 25
Rth = 7 Mbps 1 0.11 1
NRB = 25
Rth = 10 Mbps 1 0.11 1
NRB = 50
Rth = 10 Mbps 1 0.11 1
allows mitigating interference and a joint selection of suitable RBs in all FAPs, this
approach disregards certain scenarios where fading is constructive with other FAPs,
leading occasionally to better channels when an FUE is served by the FAP of an-
other apartment. With the proposed green method, this constraint is relaxed since
the purpose is to offload FUEs in order to switch FAPs off. Furthermore, switching
off certain FAPs for energy efficiency has the desirable side effect of reducing the
interference in the network, due to shutting down some (or in the simulated scenario,
most) of the transmitters. Indeed, Algorithm 2 starts from an initial implementation
of Algorithm 1, followed by an enhancement operation consisting of FAP switch off
in order to reduce the energy consumption in the network.
7.8 Conclusions
In this chapter, femtocell networks designed for supporting IoT traffic were studied.
Radio resource management and green operation in LTE and beyond (5G) femtocell
networks with centralized control was investigated. The studied scenario consisted
of an integrated wired/wireless system, where the femtocell access points are con-
22 Elias Yaacoub
Table 7.5 Normalized Utility
Centralized Green Selfish
Centralized
NRB = 15
Rth = 2 Mbps 0.90 1.0 0.79
NRB = 15
Rth = 5 Mbps 0.97 1.0 0.62
NRB = 15
Rth = 7 Mbps 0.97 0.92 0.53
NRB = 15
Rth = 10 Mbps 0.90 0.76 0.44
NRB = 25
Rth = 5 Mbps 0.98 1 0.72
NRB = 25
Rth = 7 Mbps 0.98 1 0.65
NRB = 25
Rth = 10 Mbps 0.98 0.99 0.56
NRB = 50
Rth = 10 Mbps 0.99 1 0.70
trolled by a single entity. This permits performing joint radio resource management
in a centralized and controlled way in order to enhance the quality of service perfor-
mance for all users in the networks. It also allows an energy efficient operation of the
network by switching off redundant femtocells whenever possible. Two algorithms
were proposed and analyzed. The first one is a utility maximizing radio resource
management algorithm. It was used to maximize different utility functions leading
to different target objectives in terms of network sum-rate, fairness, and enhancing
the worst-case performance in the network. The second algorithm is FAP switch off
algorithm, implemented at the central controller. The joint wired/wireless resource
management approach was compared to the distributed resource management case,
where each femtocell acts as an independent wireless network unaware of the chan-
nel and interference conditions with the other cells. The integrated wired/wireless
approach led to significant gains compared to the wireless only case, and the perfor-
mance tradeoffs between the various utility functions were analyzed and assessed.
The results of the green algorithm showed significant energy savings while satisfy-
ing QoS requirements.
7 Green 5G Femtocells for Supporting Indoor Generated IoT Traffic 23
References
1. Y.-K. Chen, “Challenges and opportunities of internet of things”, in Proc. of the Asia and
South Pacific Design Automation Conference (ASP-DAC), pp. 383–388, Sydney, Australia,
January-February 2012.
2. E. Yaacoub and A. Abu-Dayya, “Automatic Meter Reading in the Smart Grid Using Con-
tention Based Random Access over the Free Cellular Spectrum”, Computer Networks (Else-
vier), Vol. 59, pp. 171–183, February 2014.
3. E. Yaacoub and A. Kadri, “LTE Radio Resource Management for Real-Time Smart Meter
Reading in the Smart Grid”, IEEE ICC 2015 - Workshop on Green Communications and Net-
works with Energy Harvesting, Smart Grids, and Renewable Energies, London, UK, June
2015.
4. J. Lloret, A. Canovas, S. Sendra, and L. Parra, “A Smart Communication Architecture for
Ambient Assisted Living”, IEEE Communications Magazine, vol. 53, no. 1, pp.26–33, January
2015.
5. L. Mainetti, L. Patrono, and A. Vilei, “Evolution of wireless sensor networks towards the
internet of things: A survey”, in Proc. of the International Conference on Software, Telecom-
munications and Computer Networks (SoftCOM), pp. 1–6, Split, Croatia, September 2011.
6. I. Bisio, F. Lavagetto, M. Marchese, and A. Sciarrone, “Smartphone-Centric Ambient Assisted
Living Platform for Patients Suffering from Co-Morbidities Monitoring”, IEEE Communica-
tions Magazine, vol. 53, no. 1, pp.34–41, January 2015.
7. V. Chandrasekkhar, J. G. Andrews, and A. Gatherer, “Femtocell Networks: A Survey”, IEEE
Communications Magazine, vol. 46, no. 9, pp. 59–67, 2008.
8. D. Knisely, T. Yoshizawa, and F. Favichia, “Standardization of Femtocells in 3GPP”, IEEE
Communications Magazine, vol. 47, no. 9, pp. 68–75, 2009.
9. J. G. Andrews, H. Claussen, M. Dohler, S. Rangan, and M. C. Reed, “Femtocells: Past,
Present, and Future”, IEEE Journal on Selected Areas in Communications, vol. 30, no. 3,
pp. 497–508, 2012.
10. V. Chandrasekhar, M. Kountouris, and J. G. Andrews, “Coverage in Multi-Antenna Two-Tier
Networks”, IEEE Transactions on Wireless Communications, vol. 8, no. 10, pp. 5314–5327,
2009.
11. F. Pantisano, M. Bennis, W. Saad, and M. Debbah, “Spectrum Leasing as an Incentive towards
Uplink Macrocell and Femtocell Cooperation”, IEEE Journal on Selected Areas in Communi-
cations, vol. 30, no. 3, pp. 617–630, 2012.
12. C. Gussen, V. Belmega, and M. Debbah, “Pricing and Bandwidth Allocation Problems in
Wireless Multi-Tier Networks”, in Proc. of Asilomar Conference on Signals Systems and Com-
puters, pp. 1633–1637, 2011.
13. S.-Y. Lien, C.-C. Tseng, K.-C. Chen, and C.-W. Su, “Cognitive Radio Resource Management
for QoS Guarantees in Autonomous Femtocell Networks”, in Proc. of the IEEE International
Conference on Communications, (ICC 2010), pp. 1–6, 2010.
14. S. Hong, C.-Y. Oh, and T.-J. Lee, “Resource Allocation Method Using Channel Sensing and
Resource Reuse for Cognitive Femtocells”, International Journal of Information and Elec-
tronics Engineering, vol. 3, no. 3, pp. 309–312, 2013.
15. M. A. Abdelmonem, M. Nafie, M. H. Ismail, and M. S., El-Soudani, “Optimized Spectrum
Sensing Algorithms for Cognitive LTE Femtocells”, EURASIP Journal on Wireless Commu-
nications and Networking vol. 2012, no. 6, 19 pages (Open Access), 2012.
16. 3rd Generation Partnership Project (3GPP), 3GPP TS 36.211 3GPP TSG RAN Evolved Uni-
versal Terrestrial Radio Access (E-UTRA) Physical Channels and Modulation, version 12.2.0,
Release 12, 2014.
17. 3rd Generation Partnership Project (3GPP), 3GPP TS 36.213 3GPP TSG RAN Evolved Uni-
versal Terrestrial Radio Access (E-UTRA) Physical layer procedures, version 12.2.0, Release
12, 2014.
18. Qualcomm Inc., 3GPP TSG-RAN WG1 #72 R1-130598, Agenda item: 7.3.7, “Channel Mod-
els for D2D Deployments”, St. Julian’s, Malta, 2013.
24 Elias Yaacoub
19. X. Qiu, and K. Chawla, “On the Performance of Adaptive Modulation in Cellular Systems”,
IEEE Transactions on Communications, vol. 47, no. 6, pp. 884–895, 1999.
20. J.-Y. Le Boudec, “Rate adaptation, Congestion Control and Fairness: A Tutorial”, Tech. Re-
port, Ecole Polytechnique Federale de Lausanne (EPFL), Lausanne, Switzerland, 2008.
21. G. Song, and Y. Li, “Cross-Layer Optimization for OFDM Wireless Networks-Part I: Theoret-
ical Framework”, IEEE Transactions on Wireless Communications, vol. 4, no. 2, pp. 614–624,
2005.
22. E. Yaacoub and Z. Dawy, “Resource Allocation in Uplink OFDMA Wireless Systems: Opti-
mal Solutions and Practical Implementations”, John Wiley and Sons / IEEE press, NY, ISBN:
978-1-1180-7450-3, 2012.
23. E. Yaacoub, “Radio Resource Management in Integrated Wired/Wireless LTE Femtocell Net-
works”, in Proc. of the 12th International Conference on Wired and Wireless Internet Com-
munications, Paris, France, May 2014.